Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.tesis.lebedev.ru/en/docs/1570.pdf
Äàòà èçìåíåíèÿ: Tue May 12 19:28:47 2015
Äàòà èíäåêñèðîâàíèÿ: Mon Apr 11 05:09:25 2016
Êîäèðîâêà:

Ïîèñêîâûå ñëîâà: arp 220
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 110, A09S07, doi:10.1029/2004JA010931, 2005

CORONAS-F/SPIRIT EUV observations of October ­ November 2003 solar eruptive events in combination with SOHO/EIT data
V. V. Grechnev,1 I. M. Chertok,2 V. A. Slemzin,3 S. V. Kuzin,3 A. P. Ignat'ev,3 A. A. Pertsov,3 I. A. Zhitnik,3 J.-P. Delaboudinie e,4 and F. Auche e4 `r `r
Received 30 November 2004; revised 2 March 2005; accepted 8 April 2005; published 28 July 2005.

[1] The extraordinary solar activity of October ­ November 2003 manifested itself in

many powerful eruptive events, including large coronal mass ejections (CMEs) and extremely powerful flares. A number of major events were accompanied by practically all known phenomena of the solar activity, both local and large-scale, and caused severe space weather disturbances. We study large-scale posteruptive activity manifestations on the Sun associated with CMEs, i.e., dimmings and coronal waves, observed with extremeultraviolet telescopes, the SPIRIT on the CORONAS-F spacecraft and the EIT on the SOHO. During that period, observations with a cadence of 15 to 45 min were carried out by ° ° the SPIRIT in the 175 A and 304 A bands simultaneously. The EIT observed with 12-min ° band as well as with 6-hour cadence in the 171, 284, and 304 A bands. ° cadence in the 195 A These data complement each other both in the temporal and spectral coverage. Our analysis reveals that largest-scale dimmings covered almost the whole southern part of the Sun's visible side and exhibited homology, with one homological structure being changed to another configuration on 28 October. These structures show connections between large superactive and smaller regions that constituted a huge activity complex responsible for ° the extraordinary solar activity of that period. Coronal waves were observed at 175 A ° in some events, in areas where there were no active regions, but as well as at 195 A ° in the 175 A images they look fainter. They were not accompanied by deep, longliving dimmings. By contrast, such dimmings were observed in active regions, in their vicinity, and between them. These facts rule out the direct relation of the phenomena of longterm dimmings and coronal waves. On 18 November, a motion of an ejecta was observed at ° the solar disk as a propagation of a dark feature only in the 304 A band, which can be interpreted as an absorption in a ``cloud'' formed from material of the eruptive filament, which probably failed to become a CME core.
Citation: Grechnev, V. V., I. M. Chertok, V. A. Slemzin, S. V. Kuzin, A. P. Ignat'ev, A. A. Pertsov, I. A. Zhitnik, J.-P. Delaboudiniere, and F. Auchere (2005), CORONAS-F/SPIRIT EUV observations of October ­ November 2003 solar eruptive ` ` events in combination with SOHO/EIT data, J. Geophys. Res., 110, A09S07, doi:10.1029/2004JA010931.

1. Introduction
[2] A series of extreme events that occurred in October ­ November 2003 is a subject of comprehensive investigations of both violent solar activity and severe space weather disturbances (e.g., Veselovsky et al. [2004], Yermolaev et al. [2005], and papers of this issue). In particular, Chertok and Grechnev [2005, hereinafter referred to as Paper I] studied large-scale manifestations of the solar activity by full-disk heliograms gathered with the Extreme ultraviolet Imaging Telescope (EIT) [Delaboudinie e et al., 1995] aboard the `r
1 2 3 4

Institute of Solar-Terrestrial Physics SB RAS, Irkutsk, Russia. IZMIRAN, Troitsk, Moscow Region, Russia. Lebedev Institute of Physics, Moscow, Russia. Institut 'd Astrophysique Spatiale, Orsay, France.

Copyright 2005 by the American Geophysical Union. 0148-0227/05/2004JA010931$09.00

Solar and Heliospheric Observatory (SOHO). They analyzed extended dimmings, i.e., regions of temporarily reduced brightness (we do not apply the widely used term ``transient coronal holes'' to avoid confusion between decreased brightness and open magnetic structures) and, to a lesser extent, coronal waves (bright propagating or quasi-stationary fronts) associated with halo coronal mass ejections (CMEs) and powerful flares. Such phenomena have been studied previously by, for example, Hudson and Webb [1997], Thompson et al. [1998], Zarro et al. [1999], Gopalswamy and Thompson [2000], and Hudson and Cliver [2001]. The events under consideration took place during two passages across the visible disk of three big highly active (``superactive'') regions: AR 484 (501 at the second rotation, Carrington latitude and longitude N04, 354), AR 486/508 (S15, 283), and AR 488/507 (N08, 291). [3] In this paper, we continue the analysis of the October ­ November 2003 large-scale solar surface activity based
1 of 13

A09S07


A09S07

GRECHNEV ET AL.: SPIRIT AND EIT OBSERVATIONS

A09S07

on extreme ultraviolet (EUV) observations with the Spectrographic X-Ray Imaging Telescope-spectroheliograph (SPIRIT) [Zhitnik et al., 2002] aboard the CORONAS-F space solar observatory [Oraevsky and Sobelman, 2002; Oraevsky et al., 2003] launched in August 2001. Combining ° ° CORONAS-F/SPIRIT 175 A and 304 A and SOHO/EIT ° 195, 171, 304, and 284 A data [Dere et al., 2000], and taking advantage of the fact that observations of these two telescopes are complementary in the sense of the spectral and temporal coverage, we endeavor to obtain still more comprehensive picture of the large-scale surface activity associated with CMEs during that extraordinary period. [4] The present results and conceptions on the origination of dimmings are as follows. The coincidence of long-living dimmings observed in many events in different-temperature ° coronal lines and in the 304 A transition-region band [Chertok and Grechnev, 2003b] provides support for their origination due to plasma density depletion in the low corona [see, e.g., Thompson et al., 1998; Zarro et al., 1999; Harrison et al., 2003], sometimes even down to the transition region [e.g., Chertok and Grechnev, 2003a; Chertok et al., 2004a, 2004b]. Evacuation of plasmas is possible if coronal structures are opened or stretched into the interplanetary space [Delanne ´e and Aulanier, 1999], which is proposed by CME models [e.g., Antiochos et al., 1999; Veselovsky and Panasenco, 2002]. Direct evidence for plasma outflows has been provided by Doppler observations with Coronal Diagnostic Spectrometer on SOHO [Harra and Sterling, 2001]. Dissimilarities of some dimmings observed sometimes in different lines suggest that secondary temperature variations may also take place [Chertok and Grechnev, 2003b; Chertok et al., 2004a], which can be also expected. [5] Another type of large-scale activity manifestations is represented by coronal waves that were first found from EIT images (often referred to as ``EIT waves'' for that reason). Such waves have been observed almost exclusively in the ° 195 A channel because of the obvious reason: just this EIT channel is used for the imaging with a maximum rate of one full-disk image every 12 min in the standard CME Watch program. This rate is probably sufficient for detection of coronal waves but not for their comprehensive study. Combining images produced in full-disk observations with the two telescopes, SPIRIT and EIT, provides better sampling of an event. [6] Observations and conceptions of coronal waves have been reviewed by Zhukov and Auche e [2004]. Some `r interpretations [e.g., Delannee and Aulanier, 1999] connect ´ propagation of the waves with phenomena of dimmings. Observations of coronal waves with TRACE in 171 and ° 195 A almost simultaneously have been reported by WillsDavey and Thompson [1999]. Observations of a coronal wave in soft X rays with Yohkoh/SXT were discussed by Khan and Aurass [2002] and Narukage et al. [2002]. Recently, Zhukov and Auche e [2004] reported detection `r ° of coronal waves in the 284 A band. No direct indications of strong temperature dependence for coronal waves have been found so far. Taking advantage of nearly simultaneous observations with the SPIRIT and EIT in close, but different ° bands of 175 and 195 A, we also address this issue. [7] In this paper, we endeavor to make further progress in understanding of these known, but still not well-understood phenomena, posteruptive dimmings and coronal waves, and

their probable relation to each other. In addition, we use the fact that being shielded from energetic particles by the Earth magnetosphere due to the relatively low orbit of the CORONAS-F (altitude of 500 km), the SPIRIT supplies clean images when those acquired by the EIT are contaminated with ``snowstorm'' during/after proton events. This allows us to analyze those events in more detail.

2. Instrumentation
2.1. Telescopes and Observations [8] The SPIRIT complex was developed to study the solar activity in soft X-ray and EUV spectral bands. It contains two telescopes, the Ritchey ­ Chretien telescope (171, 195, ´ ° 284, and 304 A bands), whose optics is identical to that of the SOHO/EIT telescope, and the Herschel two-band tele° scope with the 175 and 304 A channels. In this paper, we discuss data obtained with the latter instrument. Because of a single reflection, its efficiency is higher in comparison with the Ritchey ­ Chretien telescope but its spectral band is ´ ° wider. The spectral efficiency function of the 175 A channel ° ° has a peak at 177 A and a FWHM of 13 A. Simulations with the CHIANTI code [Dere et al., 1997] showed that this spectral range is optimal for observations of various solar features and phenomena from coronal holes to flares in the temperature band of log10 T(MK) = 5.8 ­ 6.0. The major part of the flux detected is due to bright spectral lines of FeIX ­ XI (excitation temperature log10T = 5.9 ­ 6.1) with lesser contribution of cooler OIV-V (log10 T = 4.6 ­ 5.4) and hotter FeXX (log10 T = 6.9 ­ 7.5) lines. ° [9] The 304 A band of the Herschel telescope is peaked at ° ° 304 A and has a FWHM of 46 A. The major emission recorded in this channel contributes the transition-region HeII line (excitation temperature log T = 4.5 ­ 4.9), but this band also includes intense lines of ions SiXI, FeXV, FeXVI, and FeXXIV excited at the temperatures of log10 T > 6. The coronal and transition-region components can be separated using comparisons of simultaneous solar images in the 175 ° and 304 A bands. [10] In the standard mode, the EIT produces full-disk ° heliograms every 12 min in the 195 A band (CME Watch program), but only four times per day (every 6 hours near ° 0100, 0700, 1300, and 1900 UT) in 171, 284, and 304 A bands. The SPIRIT recorded full-disk heliograms with time intervals between consecutive frames from 15 to 45 min in the ° 175 A band during the first passage of the active complex ° (19 October to 3 November), and in the 175 and 304 A bands during its second passage (13 November to 2 December). The SPIRIT observations are periodically interrupted by occultations (night times) of $46 min in each of 15 orbits per day with an average period of $94 min. Thus simultaneous ° ° SPIRIT observations in 175 A and 304 A channels fill in ° 6-hour gaps in EIT observations in 171 and 304 A bands. [11] It is well known that ideal instruments do not exist; each one has its own advantages and problems. In observations of powerful events, differences between the instrumental characteristics of the two telescopes clearly show up. Thus combining observations with the two EUV telescopes, EIT and SPIRIT, we gain additional opportunities. [12] First, continuous observations carried out by the EIT ° in 195 A channel with 12-min interval enables uniform coverage of the whole time interval of interest, without

2 of 13


A09S07

GRECHNEV ET AL.: SPIRIT AND EIT OBSERVATIONS

A09S07

interruptions due to night time of a spacecraft flying in an orbit around the Earth. The EIT provides images of higher spatial resolution than the SPIRIT, but during strong flares they degrade from saturation effects, which interferes observations. On the other hand, observations with the SPIRIT do not show pronounced scattered light or saturation. Observations in close bands with two different telescopes simplify detection of instrumental contributions, when the features of interest appear dissimilar in their images. [13] Second, bright points and streaks, the so-called ``snowstorm,'' supposedly due to energetic particles, are present in EIT images during/after proton events just after the flare maximum. However, this stage of an event is important for studies of the development of dimmings. The CORONAS-F/ SPIRIT, being protected by the Earth magnetosphere against energetic particles, supplies clean images in those cases. Thus the combination of both EIT and SPIRIT images gives a unique opportunity to observe a clear picture of flaring active regions and long-living dimmings for a long time. 2.2. Data Processing and Representation [14] Dimmings and coronal waves are relatively faint features. Hence to detect them, differential methods are generally used. The whole area of a dimming can well exceed its appearance in nonsubtracted images [see, e.g., Zhukov and Auche e, 2004]. This also means that studying `r dimming structures looking at an original image, without subtraction of a preevent image, does not provide comprehensive information. [15] Following Paper I, we use fixed-base difference ``derotated'' images to study CME-associated disturbances. Such images are formed in two stages: first, the solar rotation is compensated for all images using three-dimensional rotation of the solar surface to the time of a base preevent heliogram (``derotation''), and then the same reference heliogram is subtracted from all others. To reveal dimmings, which are faint, it is important to choose an appropriate brightness scale. We use a linear scale in a limited range of both positive and negative brightness. Such a technique allows us to form and analyze both SPIRIT and EIT fixed-base difference images in all available spectral bands with various time intervals, up to several hours. [16] On the contrary, running-difference images, which are formed by subtraction from each heliogram the preceding one, represent the temporal derivative of the image cube. They emphasize changes of the brightness, location, and structure of emitting features between two subsequent heliograms, which is useful, e.g., to detect coronal waves. However, this method is not suitable for studies of long-duration phenomena like dimmings. Therefore we use in our study running difference images as a helpful representation only. [17] The results of our analysis are illustrated below for several the most spectacular eruptive events of two rotations of the active regions. Additional images and movies for those and other events can be found at the Web site http://helios. izmiran.troitsk.ru/lars/Chertok/0310_11/spirit/index.html.

region 486 (S15 E44), and then, at about 1800 UT, in the western active region 484 (N02 W38). In this section, we ° consider the former event using the SPIRIT (175 A) along ° with EIT (195 A) observations. This event included a longduration (LDE) 3B/X1.2 flare and a large CME of a partialhalo type with a bright loop front above the east limb. [19] The coronal wave in this event was observed with both EIT and SPIRIT as a faint, diffuse emitting structure at the interval from 0621 to 0646 UT. Figures 1a ­ 1c show its positions at different times. Figure 1e shows the spatial profiles along the arrows computed by integration over the width indicated with the bars at the arrow bases. The profiles are normalized to the preevent images. Note that the brightness ranges in Figures 1b and 1d are different according to the contrast of the wave and dimmings. [20] The positions of the wave in the images observed with the SPIRIT and EIT at different times correspond to the same moving feature (Figure 1f), although the wave is ° fainter in the 175 A band. Combining the images produced with both telescopes, we estimate the speed of the wave of about 190 km sþ1. It propagates in a confined angular segment across the northeastern quadrant of the solar disk from the large southern active region 486 toward the northern polar area, passing by side of the AR 487. The area where the wave propagates does not contain active regions. Figure 1d shows that no dimming remains in this area after the wave pass. ° [21] The 12-min cadence of EIT images at 195 A is known to be insufficient for a detailed study of coronal waves because in many cases they are visible in two to three frames only [e.g., Biesecker et al., 2002]. In this respect, it is ° important that the SPIRIT 175 A heliogram on 26 October ° was recorded at 0628 UT, i.e., between two EIT 195 A images at 0621 and 0633 UT. The combination of the EIT and SPIRIT observations provides more detailed information on the propagation of the coronal wave. [22] Already, initial EIT and SPIRIT fixed-base difference derotated images in Figures 1a ­ 1d show that besides the coronal wave, clearly visible dimmings (dark areas) appear around the eruption center in region 486. Moreover, starting from 0621 UT, a very long and narrow arc-like lane dimming develops to connect the large active regions 486 and 484 disposed far from each other. It stretches through the high-latitude area of the southern hemisphere, becoming increasingly pronounced with time, and at the late phase of ° the event predominates in both SPIRIT 175 A (Figure 1d) ° and EIT 195 A images (not shown). This channeled dimming is clearly visible also in several preceding and subsequent eruptive events (see below and Paper I). 3.2. Event of 28 October [23] Outstanding global disturbances on the solar disk on October 28 were associated with the major eruptive event that occurred after 0951 UT. A 4B flare took place in AR 486 (S16 E08). Its soft X-ray flux exceeded the saturation level of the GOES monitors (X17.2) near 1110 UT. An extremely high-speed (V % 2400 km/s) full-halo CME was observed in association with the event. This event also caused very strong proton fluxes and a severe geomagnetic storm (Dst % þ363 nT). [24] Figure 2 shows in the upper row (Figures 2a ­ 2c) ° fixed-base difference derotated 175 A SPIRIT images

3. Events of the First Rotation
3.1. Event of 26 October [18] Two similar powerful eruptive events occurred on 26 October: at first, at about 0600 UT in the southern active

3 of 13


A09S07

GRECHNEV ET AL.: SPIRIT AND EIT OBSERVATIONS

A09S07

Figure 1. The 26 October 2003, 0600 UT event in fixed-base difference ``derotated'' images observed ° ° with (b,d) SPIRIT at 175 A and (a,c) EIT at 195 A illustrating the coronal wave (marked with dashed ovals) and dimmings (dark areas). The arrows indicate approximate propagation direction of the wave. (e) The profiles along the arrows obtained by the integration over the width indicated by the bars at the arrow bases. (f) The positions of the profiles' peaks versus time. relative to the preevent time of 0924 UT. Analogous EIT ° 195 A images are presented in Figures 2d ­ 2f for approximately the same times. For this extremely intense event, that also produced powerful proton flux, the images acquired with EIT and SPIRIT differ. The first images after the eruption, in which dimmings are visible, were recorded by EIT at 1112 UT (Figure 2d) and by the SPIRIT 1 min later (Figure 2a). The dimming structure is difficult to see at that time, especially in the EIT image, because of very high brightness of the flaring region, which reached its maximum at 1110 UT in both channels. A wide saturation area in the EIT image covers significant part of dimmings, and real intensities in its vicinity are distorted by the high level of scattered light. By contrast, the dimmings are visible in the SPIRIT image much better. The difference between SPIRIT and EIT images can be explained by several reasons. First, as model calculations using the CHIANTI database (http:// wwwsolar.nrl.navy.mil/chianti.html) show, the flux for a ° ° medium flare in the 175 A band is less than in the 195 A band. Second, the signal in the SPIRIT detector does not reach full saturation, because of some nonlinearity at high brightness levels. Third, the optical design of the SPIRIT telescope is based on one-mirror Herschel scheme with a large distance between the mirror and the detector. Hence basically it has a lower level of scattered light than the shorter two-mirror Ritchey ­ Chretien EIT telescope. In the ´ next two images (not shown), both the SPIRIT and EIT display identical dimming structures, which look similar to those in Figures 2a and 2d. Thirty minutes after the flare maximum and later on, the dimming structure in the EIT images becomes indecipherable because of increasing number of bright points and streaks (the ``snowstorm'') due to energetic particles produced in the event. Only SPIRIT images can be used at this period to analyze dimmings. Thus using the combination of images from both telescopes, we can comprehensively study this event. ° [25] The SPIRIT 175 A fixed-base difference derotated images (Figures 2a ­ 2c) demonstrate that dimmings cover almost the whole southern half of the solar disk and stretch between active regions. In particular, a narrow dimming connects the western neighborhood of the eruption center in AR 486 with the environment of AR 484 near the west limb, in the same way as in the 26 October event. There is a transequatorial dimming connection between the eruption center and the northern region 488. The global picture of the disturbances throughout the whole southern hemisphere is emphasized with two bands of dimmings extending from two central active regions to the east limb, between AR 486 and 495 and 496 through AR 494 and between AR 488 and 487. A new ``horizontal'' dimming, located in the southern

4 of 13


A09S07

GRECHNEV ET AL.: SPIRIT AND EIT OBSERVATIONS

A09S07

Figure 2. One of the most powerful, geoeffective event of October 28, 2003 as visible in (a ­ c) SPIRIT ° ° 175 A and (d ­ f) EIT 195 A fixed-base difference images for approximately the same times. The SPIRIT images are free of large-area saturation and diffuse scattered light as well of the ``snowstorm'' caused by energetic particles. Global dimmings extend mainly throughout the southern half of the disk and a coronal wave (CW, marked with dashed oval) is observed with the EIT in the northern hemisphere (Figure 2d). polar segment (perhaps near the boundary of the polar coronal hole), forms during this event. Just this dimming predominates in subsequent eruptive events (see Paper I and section 3.4 of the present paper). [26] One more dimming (D), rather faint and extended, is detectable along the southeastern boundary of the western coronal hole in the northern hemisphere, which is obviously due to the decrease of brightness of a long chain of weakly emitting structures located there. The eastern end of this dimming is curved southward and comes finally to AR 488. The western end reaches the large active region 484, which is already close to the western limb. This faint dimming seems important in the context of the role of coronal holes in formation of dimmings [see, e.g., Hudson and Webb, 1997; Thompson et al., 1998; Zarro et al., 1999; Gopalswamy and Thompson, 2000; Hudson and Cliver, 2001; Chertok and Grechnev, 2003a, 2003b]. [27] Later on, the SPIRIT shows a dimming around AR 484 to evolve, likely due to an eruption in this region at about 1200 UT. It occurs, probably, close to the limb or behind it. Signatures of the eruption are detectable in EUV images, but it has not been registered by the LASCO, likely due to strong contamination of images with ``snowstorm.'' It is difficult to see this dimming in EIT images. ° [28] The EIT 195 A image of 1112 UT (Figure 2d) and, especially, the running-difference movie presented at the Web site specified in section 2 show a faintly bright structure (CW). In this frame, it is localized in the northern hemisphere, nearly in parallel to the equator, and extends from the central meridian toward the eastern limb. This feature hardly can be due to the scattered light, which is definitely present in this image, because of its elongated shape and location, which are considerably different from the scattered light pattern, especially pronounced beyond the southeast limb. In a few subsequent frames, when looking at them in a narrow brightness range, one can reveal weak brightenings in a wide sector northward and toward the northeastern limb from its position in Figure 2d, while the faintly bright structure between the central meridian and the eastern limb disappears. These few frames show intermittent faint dark and bright features in a large area from this latitude up to the north pole. These properties correspond to a coronal wave, but it is difficult to determine its propagation direction because the presence of coronal holes in that area complicates its appearance and motion. The scattered light in the EIT image does not permit us to judge about possible presence of the wave in the southern hemisphere. Like the wave in the event of 26 October, it is ° faintly visible in 175 A SPIRIT images but still can be detected in careful examination. We do not present it here in detail, but only mention its properties: (1) the wave is detectable far from active regions, whereas dimmings are localized in active regions, in their vicinity, and between them (except for the dimming along the boundary of the

5 of 13


A09S07

GRECHNEV ET AL.: SPIRIT AND EIT OBSERVATIONS

A09S07

° ° Figure 3. Comparison of (a,b) SPIRIT 175 A and (c,d) EIT 195 A fixed-base difference images for powerful event of 29 October 2003. Several large-scale channeled dimmings are better visible in cleaner SPIRIT images.

southern polar coronal hole); (2) no pronounced long-living dimming follow the wave; (3) the wave is fainter in the ° ° 175 A SPIRIT images than in the 195 A EIT images. [29] Note also a bright feature (BF) that propagates over a very long distance from AR 486 northwestward. This fact also underlines the large-scale character of CME-associated disturbances. 3.3. Event of 29 October [30] On the next day, 29 October, another very powerful and geoeffective event occurred in the same large superactive region AR 486 (S15 W02) from about 2037 UT on. It was a 2B/X10 long-duration flare associated with a full-halo CME similar to that of 28 October. The proton flux near the Earth also strongly enhanced in this case, and the accompanied severe geomagnetic storm achieved even greater value of Dst % þ401 nT. ° [31] The comparison of the SPIRIT 175 A and EIT ° 195 A fixed-base difference ``derotated'' images for this event (Figure 3) shows a similar picture. Again, we analyze the structure of dimmings based on SPIRIT images because, like the previous day, EIT images are saturated and covered with strong ``snowstorm.'' Nevertheless, one can conclude that SPIRIT and EIT images

° show very similar dimmings at 175 and 195 A, especially several hours after the eruption. [32] The clean SPIRIT images (Figures 3a and 3b) show that all dimmings are concentrated again in the southern hemisphere and the near-equatorial northern area. The interconnecting dimming observed in several preceding events between the eruption center in region 486 and the western region 484 is fragmentary. A conspicuous lane dimming along the northern boundary of the southern polar coronal hole, which initially formed 1 day before, becomes prevalent in this event, as well as a deep dimming going out of region 488 southeastwards. Besides, there are transequatorial dimmings between regions 486 and 488 and a double-branch dimming going from the same northern region 488 through region 487 to the eastern limb. 3.4. Large-Scale Homology of Dimmings [33] Figure 4 summarizes common properties of six firstrotation events that occurred on 23, 26, 28, and 29 October ° and 2 November as 175 A difference SPIRIT posteruptive images show. One can see that disturbances, which took place in those events, occurred in a huge activity complex constituted by three superactive large sunspot groups and some smaller active regions. Accordingly, a huge area was

6 of 13


A09S07

GRECHNEV ET AL.: SPIRIT AND EIT OBSERVATIONS

A09S07

° Figure 4. A collection of the SPIRIT 175 A fixed-base difference images for several eruptive events observed during the first rotation of the activity complex illustrates homology of large-scale dimmings in the southern hemisphere. Before 28 October (a ­ c), the arc-like lane structure HD1 connecting AR 484 and AR 486 predominated. Starting from 28 October on, this structure changed to another dominant homological dimming HD2 that extended along the northern boundary of the southern polar coronal hole.

occupied by disturbances ``highlighted'' with dimmings that connected widespread active regions with each other (some of those connections being transequatorial), and even reached the southern polar region. The most prominent in events that occurred by 28 October was the arc-like lane structure HD1, which connected AR 484 and AR 486. The dimming structures obviously visualize the magnetic connectivity between the active regions. [34] Some dimmings observed in many recurrent eruptive events exhibit pronounced similarity in their shapes and locations (which is often referred to as their ``homology''). Figure 4 shows that the arc-like lane structure HD1, which connected AR 484 and AR 486, dominated homologically by 28 October. Previously, Khan and Hudson [2000] analyzed dimmings in transequatorial loops. Chertok et al. [2004a] reported homological large-scale dimmings and coronal waves for events in 24 ­ 26 November 2000 associated with one active region. By contrast, during the period of October ­ November 2003, a huge spatial area was embraced by repetitive dimming structures. Then, starting from 28 October on, this structure changed to another dominant configuration HD2, which was extended along the northern boundary of the southern polar coronal hole. This latter structure is visible in images of 28 and 29 October and 2 November (Figures 4d ­ 4f). As shown in Paper I based on EIT data, this polar dimming also

dominated in the limb event of 4 November with the most powerful soft X-ray emission (this event was not recorded by the SPIRIT).

4. Events of the Second Rotation
[35] The significant eruptive activity of the huge activity complex went on also when it was on the invisible side of the Sun, during 6 ­ 12 November. Many large halo CMEs, which were recorded by LASCO without counterparts on the solar disk, certify this. The superactive regions 484, 486, and 488 appeared at the second rotation as ARs 501 (former 484; Carrington latitude and longitude N03, 002), 508 (former 486; S20, 286), and 507 (former 488; N10, 295). The eruptive events that occurred during the second pass of the activity complex across the solar disk were not as impressive as the first-rotation events; nevertheless, the activity was still high. The main dimming structure during the second rotation of the activity complex in November also embraced the southern half of the solar disk, as shown in Paper I. A series of geoeffective CME events occurred during 17 ­ 18 November in AR 501. 4.1. Event of 17 November [36] On 17 November, a 1N/M4.2 flare occurred from 0855 UT on, and a partial-halo CME was observed from

7 of 13


A09S07

GRECHNEV ET AL.: SPIRIT AND EIT OBSERVATIONS

A09S07

Figure 5. Fixed-base difference images of the 17 November 2003 event observed with 6-hour intervals ° ° in all four EIT bands of (a,b,d,e) 195, 171, 284, 304 A and (c,f) in